Maria Florina Balcan (CS) has been awarded an NSF CAREER Award to develop theoretical foundations of machine learning and related areas.

Balcan, assistant professor in the School of Computer Science, received the five-year, $400,000 award for her project, "Machine Learning Theory with Connections to Algorithmic Game Theory and Combinatorial Optimization." The project will develop theoretical foundations for novel machine learning paradigms, including active, interactive and semisupervised learning, that are needed to advance technology in areas ranging from computer vision to bio-surveillance. These paradigms were not addressed by previous mathematical models, and Balcan's work will lead to more efficient and effective machine learning algorithms. The project also will develop fundamental new connections between machine learning, game theory and combinatorial optimization that will aid in advancing and solving important problems in all three areas.

"Machine learning is now being used in so many different areas, and the classic models we have just don't apply so well anymore," Balcan said. "Also, at a mathematical level, the core issues that come up again and again in machine learning are closely connected to those in game theory and optimization--for instance, issues of adapting to change and to new information. Bringing these areas together can produce understanding that can advance all three."

To learn more about Balcan’s CAREER Award or to view the full abstract, visit the NSF website.